{"id":8304,"date":"2021-11-08T10:28:18","date_gmt":"2021-11-08T15:28:18","guid":{"rendered":"https:\/\/dronebotworkshop.com\/?p=8304"},"modified":"2023-04-12T07:10:27","modified_gmt":"2023-04-12T11:10:27","slug":"huskylens","status":"publish","type":"post","link":"https:\/\/dronebotworkshop.com\/huskylens\/","title":{"rendered":"Huskylens &#8211; AI Machine Vision Sensor"},"content":{"rendered":"\n<p><img loading=\"lazy\" decoding=\"async\" class=\"wp-image-9571 size-full alignnone\" title=\"Download PDF\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2023\/04\/dbws-pdfdown.png?resize=64%2C64&#038;ssl=1\" alt=\"Download PDF\" width=\"64\" height=\"64\" data-recalc-dims=\"1\" \/> <a href=\"#Parts_List\"><img loading=\"lazy\" decoding=\"async\" class=\"alignnone wp-image-9563 size-full\" title=\"Parts List\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2023\/04\/dbws-parts.png?resize=64%2C64&#038;ssl=1\" alt=\"Parts List\" width=\"64\" height=\"64\" data-recalc-dims=\"1\" \/><\/a> <a href=\"https:\/\/youtu.be\/E140gPLPz4A\" target=\"_blank\" rel=\"noopener\"><img loading=\"lazy\" decoding=\"async\" class=\"alignnone wp-image-9565 size-full\" title=\"View on YouTube\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2023\/04\/dbws-youtube.png?resize=64%2C64&#038;ssl=1\" alt=\"View on YouTube\" width=\"64\" height=\"64\" data-recalc-dims=\"1\" \/><\/a> <a href=\"https:\/\/s3.amazonaws.com\/download.dronebotworkshop.com\/HuskyLens+-+DroneBot+Workshop.zip\"><img loading=\"lazy\" decoding=\"async\" class=\"alignnone wp-image-9569 size-full\" title=\"Download Code\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2023\/04\/dbws-codedown.png?resize=64%2C64&#038;ssl=1\" alt=\"Download Code\" width=\"64\" height=\"64\" data-recalc-dims=\"1\" \/><\/a><\/p>\n<p><span style=\"font-weight: 400;\">Machine vision is a fascinating subject that, until recently, required a lot of expensive computer \u201chorsepower\u201d, as well as an understanding of complex algorithms and coding.\u00a0 But today\u2019s new generation of powerful microcontrollers has changed all of that, and there are now several low-cost machine vision solutions that are perfect for hobbyists and experimenters.<\/span><\/p>\n<div class=\"youtube-player\" data-id=\"E140gPLPz4A\"><\/div>\n<p><span style=\"font-weight: 400;\">I have a few of those solutions to share with you and the first one, which is presented here today, is a very cool product from DFRobot called the HuskyLens.<\/span><\/p>\n<h2><span style=\"font-weight: 400;\">Introduction<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">We have already looked at a machine vision product, the PixyCam. This little board has an onboard camera and was capable of simple image recognition and line following, among other features.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8264\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/HuskyLens.jpg?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/HuskyLens.jpg?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/HuskyLens.jpg?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">The board we are going to examine today, the HuskyLens, is similar to the PixyCam, but much more powerful. It has both a camera and a small color display and can be used as a stand-alone product or as a peripheral for a microcontroller or microcomputer.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">This device is ideal for a number of image recognition and tracking applications.<\/span><\/p>\n<h2><span style=\"font-weight: 400;\">HuskyLens<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">HuskyLens was brought to life as a <\/span><a href=\"https:\/\/www.kickstarter.com\/projects\/1371216747\/huskylens-an-ai-camera-click-learn-and-play\"><span style=\"font-weight: 400;\">Kickstarter campaign<\/span><\/a><span style=\"font-weight: 400;\"> in 2020. It is now for sale on the DFRobot website, as well as through many electronics distributors (I purchased mine at DigiKey).<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The HuskyLens is available in two versions, the only difference between the two is the video sensor, which determines the resolution of the camera:<\/span><\/p>\n<ul>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><b>HuskyLens<\/b><span style=\"font-weight: 400;\"> &#8211; OV2640 (2 MP camera)<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><b>HuskyLens PRO<\/b><span style=\"font-weight: 400;\"> &#8211; OV5640 (5 MP camera)<\/span><\/li>\n<\/ul>\n<p><span style=\"font-weight: 400;\">The HuskyLens that I am using is the 2 MP camera model.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Features &amp; Specifications<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">Both models of the HuskyLens have similar specifications. The devices can be used with either 3.3-volt or 5-volt power. At 3.3-volts they consume an average of 320mA, whereas operation under 5-volts consumes 230ma (both figures quoted in face recognition mode).<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The display on the HuskyLens is a 2-inch IPS screen with a resolution of 320 x 240 pixels. The module itself measures 52mm x 44.5mm.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The HuskyLens can communicate with your microcontroller or microcomputer using either I2C or a serial UART connection.\u00a0 The 4-pin JST connector on the board is used for both protocols, and the device can either be set for one or can auto-detect the protocol on its own.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">DFRobot provides a JST cable with four female DuPont pins on the other end, to simplify connecting the HuskyLens to a microcontroller or microcomputer.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Power for the HuskyLens can be provided through either the two power pins on the 4-pin JST connector or with a microUSB connector. If both power sources are connected, the USB power will be used.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The HuskyLens also has a microSD card slot. The microSD card can be used to store images, it can also store<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The HuskyLens comes with a silicone cover (with \u201cHusky ears\u201d) and a mounting bracket.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Kendryte K210<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">The heart of the HuskyLens is the powerful <\/span><a href=\"https:\/\/canaan.io\/product\/kendryteai\"><span style=\"font-weight: 400;\">Kendryte K210<\/span><\/a><span style=\"font-weight: 400;\"> microcontroller.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">This chip has a 64bit RISC-V Dual Core CPU. It features Kendryte\u2019s self-developed \u201cKPU\u201d, a high-performance neural network hardware accelerator that can perform convolutional neural network operations.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The chip also has a high-performance microphone array audio processor that can simultaneously process the inputs from up to eight microphones.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The K210 is highly suited for image recognition and speech recognition applications.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Controlling the HuskyLens<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">There are two controls on the HuskyLens that can be used in stand-alone mode. The device can also be controlled programmatically using the attached microcontroller or microcomputer.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">One of the controls is a pushbutton, the other is a \u201cdial\u201d that can also function as a pushbutton.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">These buttons both have multiple uses, which will be described in detail further in this article.<\/span><\/p>\n<h2><span style=\"font-weight: 400;\">HuskyLens Firmware Update<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">Before you start using your HuskyLens you should check the current firmware version. If it is less than 0.5.1Norm then you should update the firmware.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">You\u2019ll need to connect the HuskyLens to a computer to perform the firmware update. Although you can find instructions for Linux, Mac, and Windows on the <\/span><a href=\"https:\/\/wiki.dfrobot.com\/HUSKYLENS_V1.0_SKU_SEN0305_SEN0336\"><span style=\"font-weight: 400;\">DFRobot HuskyLens Wiki<\/span><\/a><span style=\"font-weight: 400;\">, you are going to find it much easier to use Windows for the firmware update.\u00a0 That is because there is a GUI updater available for Windows, whereas the other operating systems require you to use the command line.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Assuming you are using windows, you\u2019ll need to download two things from the <\/span><a href=\"https:\/\/github.com\/HuskyLens\/HUSKYLENSUploader\"><span style=\"font-weight: 400;\">HuskyLens GitHub page<\/span><\/a><span style=\"font-weight: 400;\">:<\/span><\/p>\n<ul>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">The <\/span><a href=\"https:\/\/raw.githubusercontent.com\/HuskyLens\/HUSKYLENSUploader\/master\/HUSKYLENS%20Uploader-V2.1.zip\"><span style=\"font-weight: 400;\">Windows Firmware Updater<\/span><\/a><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">The firmware itself &#8211; this is a <\/span><a href=\"https:\/\/raw.githubusercontent.com\/HuskyLens\/HUSKYLENSUploader\/master\/HUSKYLENSWithModelV0.5.1aNorm.kfpkg\"><span style=\"font-weight: 400;\">link to version 0.5.1Norm<\/span><\/a><\/li>\n<\/ul>\n<p><span style=\"font-weight: 400;\">You will also need to download a USB to UART bridge, which <\/span><a href=\"https:\/\/www.silabs.com\/developers\/usb-to-uart-bridge-vcp-drivers\"><span style=\"font-weight: 400;\">you can get from Silicon Labs<\/span><\/a><span style=\"font-weight: 400;\">.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The procedure for performing the firmware update under Microsoft Windows is as follows:<\/span><\/p>\n<ul>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Start with the HuskyLens disconnected from the computer.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Download the required files from the links above and unzip them.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Run the USB to UART bridge.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Plug the HuskyLens into a USB port<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Check your Device Manager and confirm that you see a \u201cSilicon Labs\u201d device connected to a USB port.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Install the HuskyLens Updater.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Run the Updater. When prompted to choose a firmware file, select the one you have downloaded.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">The upgrade process will begin<\/span><\/li>\n<\/ul>\n<p><span style=\"font-weight: 400;\">The firmware update can take at least five minutes, the graph on the updater will go from 0 to 100 several times during the process. Just be patient, and your HuskyLens update will succeed.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Once the update is complete, check the HuskyLens firmware version again. It should show the updated version now.<\/span><\/p>\n<h2><span style=\"font-weight: 400;\">Modes of Operation<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">The HuskyLens has several modes of operation:<\/span><\/p>\n<ul>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Face Recognition<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Object Tracking<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Object Recognition<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Line Tracking<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Color Recognition<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Tag Recognition<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Object Classification<\/span><\/li>\n<\/ul>\n<p><span style=\"font-weight: 400;\">Let\u2019s examine each of these modes in more detail.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Face Recognition<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">In this mode, the HuskyLens can detect faces, assign an identifier to them, recognize them afterward, and track their movements.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The device is capable of learning faces individually or in groups.<\/span><\/p>\n<h4><span style=\"font-weight: 400;\">Learning a Face<\/span><\/h4>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8263\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/face-recognition-single-face.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/face-recognition-single-face.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/face-recognition-single-face.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">We\u2019ll begin by placing the HuskyLens in Face Recognition mode, which is accomplished by turning the selector dial at the top until \u201cFace Recognition\u201d appears on the menu at the top of the screen.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The screen will have a \u201ccross\u201d cursor at the center. Point that at a face you want to detect. When the HuskyLens detects the shape of a face, it will place a box around it, with the word \u201cFace\u201d at the top.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Now briefly press the <\/span><i><span style=\"font-weight: 400;\">Learning<\/span><\/i><span style=\"font-weight: 400;\"> button (the pushbutton on the top right) and the HuskyLens will learn the face and save it with a label of \u201cFace:ID0\u201d for the first face.\u00a0 Subsequent faces will be labeled \u201cFace:ID1\u201d, \u201cFace:ID2\u201d etc.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Although the face has been saved, it will only be recognized when viewed at the same angle as the sample face.\u00a0 In the \u201creal world\u201d this will lead to poor performance.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">In order to better recognize a three-dimensional face or a moving face in a video, you can train it for a longer period of time.<\/span><\/p>\n<h4><span style=\"font-weight: 400;\">Deleting a Learned Face<\/span><\/h4>\n<p><span style=\"font-weight: 400;\">By default, you can only learn one face. Other faces will be recognized as \u201cface\u201d, but they can\u2019t be learned.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">We can set the HuskyLens into Multiple face mode to recognize more than one face. However, in order to do that, we will first need to delete the face we have already learned.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">In order to delete a face, or anything you have learned, you do the following:<\/span><\/p>\n<ul>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Get the face you want to delete into the screen where it is recognized. It does not need to be in the center, as long as HuskyLens can identify it, then it can be anywhere on the screen.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Click the Learning key. A menu will pop up, asking if you would like to forget the identified object, in this case the face.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Respond Yes using the wheel control.\u00a0 The face will no longer be recognized.<\/span><\/li>\n<\/ul>\n<h4><span style=\"font-weight: 400;\">Learning a group of faces<\/span><\/h4>\n<p><span style=\"font-weight: 400;\">Now that we have a clean slate, we can set up the HuskyLens to recognize multiple faces.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8262\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/face-recognition-multiple-faces.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/face-recognition-multiple-faces.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/face-recognition-multiple-faces.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">The process of setting up the HuskyLens for learning multiple objects is identical for all HuskyLens modes, not just Face Tracking:<\/span><\/p>\n<ul>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Enter the <\/span><i><span style=\"font-weight: 400;\">Face Tracking<\/span><\/i><span style=\"font-weight: 400;\"> mode using the wheel control and observe the top of the display.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Press down on the wheel &#8211; you should see a new display at the bottom of the screen.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Now use this new screen to navigate to <\/span><i><span style=\"font-weight: 400;\">Learn Multiple<\/span><\/i><span style=\"font-weight: 400;\">.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Press down on the wheel &#8211; you should see a toggle switch display on the screen.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Use the wheel to toggle the switch, it is ON when set to the right and colored blue. OFF is when it is to the left and colored white.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Set the switch ON for learning multiple items.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Use the wheel to navigate back to <\/span><i><span style=\"font-weight: 400;\">Save &amp; Return<\/span><\/i><span style=\"font-weight: 400;\">.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Press down on the wheel on <\/span><i><span style=\"font-weight: 400;\">Save &amp; Return<\/span><\/i><span style=\"font-weight: 400;\">. Confirm by pressing down again.<\/span><\/li>\n<\/ul>\n<p><span style=\"font-weight: 400;\">Now the HuskyLens is set up to track multiple faces.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Track the first face as you did for single faces, aim the crosshair mark at the target face, and press down on the Learning key. Hold it down a whale to capture the face from different angles. Then release it, and the first face will be assigned an ID number.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8261\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/face-recognition-learn-second-face.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/face-recognition-learn-second-face.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/face-recognition-learn-second-face.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">You will get a message asking if you want to learn another face. If you do, then click on the Learning key once to confirm.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">If you confirm that you want to learn another face, you\u2019ll get the crosshairs again, which you can aim at another face. Once again, press the Learning key to learn the face.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Continue this process to memorize as many faces as you wish.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Once you are done learning faces, you can either press down upon the navigation wheel or just wait until the message time out.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The HuskyLens can now identify multiple faces.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Object Tracking<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">In Object Tracking mode, you can train the HuskyLens to learn one object. The object can be just about anything, and even gestures count as objects.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">You can enter Object Tracking mode by navigating with the selector wheel until <\/span><i><span style=\"font-weight: 400;\">Object Tracking <\/span><\/i><span style=\"font-weight: 400;\">is displayed on the top panel. Press down upon the selector wheel to select <\/span><i><span style=\"font-weight: 400;\">Object Tracking<\/span><\/i><span style=\"font-weight: 400;\">.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">In order to learn a new object, you need to place the HuskyLens into learn mode. Press down upon the selector wheel and another menu will appear at the bottom of the screen.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8299\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-tracking-learn-enable.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-tracking-learn-enable.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-tracking-learn-enable.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">Navigate to<\/span><i><span style=\"font-weight: 400;\"> Learn Enable<\/span><\/i><span style=\"font-weight: 400;\">, which is another switch control. Use the selector wheel to activate Learn Enable.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Perform the same operation with the <\/span><i><span style=\"font-weight: 400;\">Autosave<\/span><\/i><span style=\"font-weight: 400;\"> option, which you also want to be activated.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Now navigate back to <\/span><i><span style=\"font-weight: 400;\">Save &amp; Return<\/span><\/i><span style=\"font-weight: 400;\"> and press the selector wheel twice to accept and confirm your changes.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8298\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-tracking-autosave.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-tracking-autosave.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-tracking-autosave.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">Now you&#8217;re ready to start learning an object. Aim the HuskyLens at the target object and press down and hold the Learning key. Keep the key pressed as you scan the object from several angles.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">After you release the Learning key, the HuskyLens will recognize the object and will be able to differentiate it from similar objects.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8300\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-tracking-learn-object.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-tracking-learn-object.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-tracking-learn-object.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">The HuskyLens will continue to learn the saved object every time that it sees it, for improved performance.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Object Recognition<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">The HuskyLens has been programmed to recognize twenty different objects, as follows:<\/span><\/p>\n<ol>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Airplane<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Bicycle<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Bird<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Boat<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Bottle<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Bus<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Car<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Cat<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Chair<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Cow<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Dog<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Horse<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Motorbike<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Person<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Plant<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Sheep<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Sofa<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Table<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Train<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Television<\/span><\/li>\n<\/ol>\n<p><span style=\"font-weight: 400;\">You just need to select <\/span><i><span style=\"font-weight: 400;\">Object Recognition<\/span><\/i><span style=\"font-weight: 400;\"> using the selector wheel and point it at your target area. If one of the preprogrammed objects appears in the HuskyLens field of view, it will be displayed and tracked. Multiple objects of different types are supported.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Here are a few samples, for more please see the accompanying video.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8295\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-recognition-people.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-recognition-people.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-recognition-people.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8297\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-recognition-sheep-cow.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-recognition-sheep-cow.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-recognition-sheep-cow.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8293\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-recognition-cat.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-recognition-cat.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-recognition-cat.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8294\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-recognition-cow.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-recognition-cow.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-recognition-cow.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8296\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-recognition-people-bikes.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-recognition-people-bikes.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-recognition-people-bikes.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">As you can see from the above samples, the HuskyLens does a pretty admirable job of object recognition.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Line Tracking<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">In Line Tracking mode, the HuskyLens can be programmed to recognize and track one or more colored lines.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Although the default mode is set to tracking multiple lines, you\u2019ll often find it easier to track single lines, so you should turn off Learn Multiple unless you really need it. You can do that in the same fashion as you did for the other modes, select the mode, press the selector switch and navigate through the lower menu to Learn Multiple, where you can change the setting with the toggle switch.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8288\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/line-tracking-led-switch.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/line-tracking-led-switch.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/line-tracking-led-switch.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">You might also want to activate the white LEDs while you are in the bottom menu, as they can assist in line tracking.\u00a0 They use a toggle switch as well.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Once you have made your changes, you can navigate to Save &amp; Return to confirm your settings and exit.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Now you can begin to learn lines. Focus the HuskyLens on the line you want to use and press the Learning key. Hold it down while you track the line from various angles.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8287\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/line-tracking-learning-line.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/line-tracking-learning-line.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/line-tracking-learning-line.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">As you hold down the Learning key, the HuskyLens will display a yellow arrow in the direction of the target line, this indicates that it is learning the line. Once you release the key, it will display the tracked line in blue.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8286\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/line-tracking-learned-blue-line.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/line-tracking-learned-blue-line.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/line-tracking-learned-blue-line.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">You will notice that it is quite good at tracking vertical lines, but not so good at horizontal ones. For most line tracking applications, this is the preferred operation.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Color Recognition<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">In Color Recognition mode, the HuskyLens can learn and distinguish several colors.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Like most of the other modes, you can set up the HuskyLens to recognize one or many colors, and the default is just one. To set it up for multiple colors, dial up <\/span><i><span style=\"font-weight: 400;\">Learn Multiple<\/span><\/i><span style=\"font-weight: 400;\"> on the bottom menu and use the toggle switch to select this.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Now aim the HuskyLens at the first color and pres the Learning key down. If the color is on a flat surface, then you just need to give the key a brief press.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8258\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/color-recognition-learn-colors.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/color-recognition-learn-colors.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/color-recognition-learn-colors.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">After you release the key, you\u2019ll be prompted to click again to learn another color &#8211; well technically you\u2019re prompted to \u201cClink\u201d again, but I think you know what they mean!\u00a0 Continue to memorize as many colors as you need.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8257\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/color-recognition-learn-additional-colors.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/color-recognition-learn-additional-colors.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/color-recognition-learn-additional-colors.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">After you finish learning colors, press down on the selector dial to exit learning mode.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8259\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/color-recognition-track-multiple-colors.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/color-recognition-track-multiple-colors.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/color-recognition-track-multiple-colors.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">Now the HuskyLens will identify the memorized colors whenever it sees them.\u00a0 Note that lighting conditions can play a big part in the accuracy of this prowess, so if you\u2019re having trouble getting reliable results, try changing or moving the light source.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Tag Recognition<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">AprilTags are a creation of the <\/span><a href=\"https:\/\/april.eecs.umich.edu\/software\/apriltag\"><span style=\"font-weight: 400;\">University of Michigan April Robotics laboratory<\/span><\/a><span style=\"font-weight: 400;\">. They can be used to identify areas, objects, or actions and serve as calibration targets for robots.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The HuskyLens can learn and identify multiple AprilTags. It does not support any other type of tagging system, at least it doesn\u2019t in the current firmware release.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Learning AprilTags is similar to learning colors. You will want to be in Learn Multiple mode in most cases, so select that in the usual fashion.\u00a0 Then aim the HuskyLens at the first AprilTag you want to learn and press the Learning key. You only need to be brief, as a tag is a simple two-dimensional object.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8302\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/tag-recognition-learn-multiple-tags.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/tag-recognition-learn-multiple-tags.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/tag-recognition-learn-multiple-tags.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">Continue to learn as many tags as you wish.\u00a0 When you are done, press down on the selector wheel to end the learning process.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8301\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/tag-recognition-different-angle.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/tag-recognition-different-angle.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/tag-recognition-different-angle.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">Now the HuskyLens will detect the learned AprilTags. Note that it is capable of detecting them at any angle, not just the one you trained it in.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Object Classification<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">In Object Classification mode, the HuskyLens will allow you to classify several objects. For example, you could categorize different types of fruit like apples, oranges, lemons, etc.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">As usual, you\u2019ll want to be in Learn Multiple mode to work with multiple objects, so select it in the standard fashion.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8292\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-classification-learn-object.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-classification-learn-object.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-classification-learn-object.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">Once you have done that, aim the HuskyLens at the first object that you wish to categorize and hold down the Learning key. Continue to hold down the key and learn the object from every angle that you can.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">After you have finished learning the object, release the key. As with the other modes, you\u2019ll be prompted to click again to learn another object.\u00a0 Continue and recognize as many objects as you like.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8291\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-clasification-recognize-object.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-clasification-recognize-object.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/object-clasification-recognize-object.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">When you have finished learning objects,\u00a0 press down upon the selector wheel to end the training process.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The HuskyLens will now recognize the objects you have classified.<\/span><\/p>\n<h2><span style=\"font-weight: 400;\">HuskyLens with Arduino<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">While the HuskyLens is a pretty impressive performer in its own right, the real excitement comes with using it as an intelligent \u201cedge computing\u201d device for a microcontroller.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">As you recall from the earlier discussion about the HuskyLens specifications, this device can communicate with a host microcontroller in two fashions &#8211; Serial UART and I2C.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">We are going to interface an Arduino Uno with the HuskyLens using both methods, starting with the Serial UART.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">HuskyLens Hookup &#8211; UART Mode<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">In Serial UART mode, the HuskyLens is connected to the Arduino Uno as follows:<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8266\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/huskylens-arduino-uart.jpg?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/huskylens-arduino-uart.jpg?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/huskylens-arduino-uart.jpg?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">You can use the cable that comes with the HuskyLens, along with four short M-M jumper wires, to make the connection to the Arduino directly.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Once you have everything wired up, connect the Arduino to your computer. But before you start programming, there is one more task you need to take care of.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Installing the HuskyLens Library<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">DFRobot has provided a library for the HuskyLens, it really makes working with this little wonder very simple. However, the Library is not on the Arduino Library Manager list, nor is it available in an easy to import ZIP file. Instead, you will need to install it manually.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">It\u2019s probably best NOT to have the Arduino IDE open while you do this (although you won\u2019t hurt anything if you do), so if it\u2019s open you may wish to close it before proceeding.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">First, you\u2019ll need to obtain the library from the <\/span><a href=\"https:\/\/github.com\/HuskyLens\/HUSKYLENSArduino\"><span style=\"font-weight: 400;\">DFRobot HuskyLens Arduino GitHub page<\/span><\/a><span style=\"font-weight: 400;\">. On that page you\u2019ll see all the resources for the HuskyLens Arduino API.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">You\u2019ll also see a green <\/span><i><span style=\"font-weight: 400;\">Code<\/span><\/i><span style=\"font-weight: 400;\"> button with an arrow in it. Click on the arrow, and a sub-menu will drop down. One of the selections in the sub-menu is <\/span><i><span style=\"font-weight: 400;\">Download ZIP,<\/span><\/i><span style=\"font-weight: 400;\"> which is what you want to do.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">This ZIP file contains the library, but the directory structure is incorrect for the Arduino IDEs <\/span><i><span style=\"font-weight: 400;\">Import Library<\/span><\/i><span style=\"font-weight: 400;\"> function.\u00a0 So we will need to do the installation ourselves.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">After you download the ZIP file, open it up with your archive tool.\u00a0 Now navigate within the unzipped folder, you will see the root folder is called <\/span><i><span style=\"font-weight: 400;\">HUSKYLENSArduinoMaster<\/span><\/i><span style=\"font-weight: 400;\">. Inside that folder you should see three files plus another folder called <\/span><i><span style=\"font-weight: 400;\">HUSKYLENS<\/span><\/i><span style=\"font-weight: 400;\">.\u00a0\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Highlight the <\/span><i><span style=\"font-weight: 400;\">HUSKYLENS<\/span><\/i><span style=\"font-weight: 400;\"> folder and extract it to your <\/span><i><span style=\"font-weight: 400;\">Arduino libraries<\/span><\/i><span style=\"font-weight: 400;\"> folder.\u00a0 You\u2019ll find the <\/span><i><span style=\"font-weight: 400;\">libraries<\/span><\/i><span style=\"font-weight: 400;\"> folder in the same <\/span><i><span style=\"font-weight: 400;\">arduino<\/span><\/i><span style=\"font-weight: 400;\"> folder that your Arduino sketch folders are stored in, which in most operating systems is in your <\/span><i><span style=\"font-weight: 400;\">Home<\/span><\/i><span style=\"font-weight: 400;\"> or <\/span><i><span style=\"font-weight: 400;\">User<\/span><\/i><span style=\"font-weight: 400;\"> folder.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Once you have finished extracting the files, you can close your extraction tool, then open the Arduino IDE. Navigate to the <\/span><i><span style=\"font-weight: 400;\">File\/Examples<\/span><\/i><span style=\"font-weight: 400;\"> sub-menu and scroll down into the section of <\/span><i><span style=\"font-weight: 400;\">Examples from Custom Libraries<\/span><\/i><span style=\"font-weight: 400;\">. You should see a <\/span><i><span style=\"font-weight: 400;\">HUSKYLENS<\/span><\/i><span style=\"font-weight: 400;\"> entry, which contains some sample code.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Let\u2019s run some of that code right now.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">HuskyLens Arduino Demo &#8211; UART Mode<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">The first example sketch that we will look at is the one titled <\/span><i><span style=\"font-weight: 400;\">GETTING_STARED<\/span><\/i><span style=\"font-weight: 400;\">, which probably means \u201cgetting started\u201d. It\u2019s one of the example sketches found in the HUSKYLENS examples folder we just looked for.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">This sketch illustrates how to connect to the HuskyLens and read data from it. The data can be face, object or line data. The results will be displayed in the serial monitor.<\/span><\/p>\n<pre class=\"lang:c++ decode:true \" title=\"HuskyLens Getting Started\">\/***************************************************\r\n HUSKYLENS An Easy-to-use AI Machine Vision Sensor\r\n &lt;https:\/\/www.dfrobot.com\/product-1922.html&gt;\r\n \r\n ***************************************************\r\n This example shows the basic function of library for HUSKYLENS via Serial.\r\n \r\n Created 2020-03-13\r\n By [Angelo qiao](Angelo.qiao@dfrobot.com)\r\n \r\n GNU Lesser General Public License.\r\n See &lt;http:\/\/www.gnu.org\/licenses\/&gt; for details.\r\n All above must be included in any redistribution\r\n ****************************************************\/\r\n\r\n\/***********Notice and Trouble shooting***************\r\n 1.Connection and Diagram can be found here\r\n &lt;https:\/\/wiki.dfrobot.com\/HUSKYLENS_V1.0_SKU_SEN0305_SEN0336#target_23&gt;\r\n 2.This code is tested on Arduino Uno, Leonardo, Mega boards.\r\n ****************************************************\/\r\n\r\n#include \"HUSKYLENS.h\"\r\n#include \"SoftwareSerial.h\"\r\n\r\nHUSKYLENS huskylens;\r\nSoftwareSerial mySerial(10, 11); \/\/ RX, TX\r\n\/\/HUSKYLENS green line &gt;&gt; Pin 10; blue line &gt;&gt; Pin 11\r\nvoid printResult(HUSKYLENSResult result);\r\n\r\nvoid setup() {\r\n    Serial.begin(115200);\r\n    mySerial.begin(9600);\r\n    while (!huskylens.begin(mySerial))\r\n    {\r\n        Serial.println(F(\"Begin failed!\"));\r\n        Serial.println(F(\"1.Please recheck the \\\"Protocol Type\\\" in HUSKYLENS (General Settings&gt;&gt;Protocol Type&gt;&gt;Serial 9600)\"));\r\n        Serial.println(F(\"2.Please recheck the connection.\"));\r\n        delay(100);\r\n    }\r\n}\r\n\r\nvoid loop() {\r\n    if (!huskylens.request()) Serial.println(F(\"Fail to request data from HUSKYLENS, recheck the connection!\"));\r\n    else if(!huskylens.isLearned()) Serial.println(F(\"Nothing learned, press learn button on HUSKYLENS to learn one!\"));\r\n    else if(!huskylens.available()) Serial.println(F(\"No block or arrow appears on the screen!\"));\r\n    else\r\n    {\r\n        Serial.println(F(\"###########\"));\r\n        while (huskylens.available())\r\n        {\r\n            HUSKYLENSResult result = huskylens.read();\r\n            printResult(result);\r\n        }    \r\n    }\r\n}\r\n\r\nvoid printResult(HUSKYLENSResult result){\r\n    if (result.command == COMMAND_RETURN_BLOCK){\r\n        Serial.println(String()+F(\"Block:xCenter=\")+result.xCenter+F(\",yCenter=\")+result.yCenter+F(\",width=\")+result.width+F(\",height=\")+result.height+F(\",ID=\")+result.ID);\r\n    }\r\n    else if (result.command == COMMAND_RETURN_ARROW){\r\n        Serial.println(String()+F(\"Arrow:xOrigin=\")+result.xOrigin+F(\",yOrigin=\")+result.yOrigin+F(\",xTarget=\")+result.xTarget+F(\",yTarget=\")+result.yTarget+F(\",ID=\")+result.ID);\r\n    }\r\n    else{\r\n        Serial.println(\"Object unknown!\");\r\n    }\r\n}\r\n<\/pre>\n<p><span style=\"font-weight: 400;\">The sketch starts by including the required libraries, the HuskyLens library, and the <\/span><a href=\"https:\/\/www.arduino.cc\/en\/Reference\/softwareSerial\"><span style=\"font-weight: 400;\">SoftwareSerial library<\/span><\/a><span style=\"font-weight: 400;\">. We will be requiring SoftwareSerial as the Arduino Uno we are using only has one serial port.\u00a0 The library will let us define a second one using software.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">We then create objects for both the HuskyLens and our software-created serial port. Note that we pass the Arduino pins that we have used to connect to the HuskyLens to the SoftwareSerial object. If you\u2019d like to use different pins for the connection, you would need to change these numbers.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">In the Setup, we initialize both the hardware and software serial connections, the hardware one being for our Serial Monitor.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">We then connect the HuskyLens to the SoftwareSerial connection, displaying an error message on the Serial Monitor if the connection fails.\u00a0 If it does fail you\u2019ll want to check the wiring from the HuskyLens to the Arduino.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Assuming that we establish a connection, we jump into the Loop.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">In the Loop, we start by checking several things:<\/span><\/p>\n<ul>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Do we have an established connection to the HuskyLens?<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Has the HuskyLens learned anything (i.e. an object or line)?<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Is one of those learned objects detected on the screen right now?<\/span><\/li>\n<\/ul>\n<p><span style=\"font-weight: 400;\">If the answer to all three questions is YES, then the code jumps into a while loop, reading the data and displaying it to the Serial Monitor.\u00a0 If the data drops off, such as when the target object goes out of view, the Loop will repeat.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The data for the Serial Monitor is formatted by a function called <\/span><i><span style=\"font-weight: 400;\">printResult<\/span><\/i><span style=\"font-weight: 400;\">. It determines if the detected item is a line or a block (such as a face or an object) and formats the text appropriately.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Load the code onto the Arduino, set the HuskyLens in the desired mode and observe the serial monitor. I tried it in Face Recognition mode and the display gave me the position of the block that detected the face, along with the face ID number.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">You can see how you could make use of this information in your own project code.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">HuskyLens Arduino Hookup &#8211; I2C Mode<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">You can also use the HuskyLens in I2C mode. In order to do this, you\u2019ll need to change your wiring slightly.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8265\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/huskylens-ardiono-i2c.jpg?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/huskylens-ardiono-i2c.jpg?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/huskylens-ardiono-i2c.jpg?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">The VCC (5V) and Ground connections remain the same. The TX (green wire) is moved to A4, and the other wire (the blue RX wire) goes to pin A5.\u00a0<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">HuskyLens Arduino Demo &#8211; I2C Mode<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">The next sketch is also included in the HuskyLens examples that were installed along with the library. It is exactly the same sketch we looked at earlier, except this time we connect to the HuskyLens via I2C.<\/span><\/p>\n<pre class=\"lang:c++ decode:true \" title=\"HuskyLens I2C\">\/***************************************************\r\n HUSKYLENS An Easy-to-use AI Machine Vision Sensor\r\n &lt;https:\/\/www.dfrobot.com\/product-1922.html&gt;\r\n \r\n ***************************************************\r\n This example shows the basic function of library for HUSKYLENS via I2c.\r\n \r\n Created 2020-03-13\r\n By [Angelo qiao](Angelo.qiao@dfrobot.com)\r\n \r\n GNU Lesser General Public License.\r\n See &lt;http:\/\/www.gnu.org\/licenses\/&gt; for details.\r\n All above must be included in any redistribution\r\n ****************************************************\/\r\n\r\n\/***********Notice and Trouble shooting***************\r\n 1.Connection and Diagram can be found here\r\n &lt;https:\/\/wiki.dfrobot.com\/HUSKYLENS_V1.0_SKU_SEN0305_SEN0336#target_23&gt;\r\n 2.This code is tested on Arduino Uno, Leonardo, Mega boards.\r\n ****************************************************\/\r\n\r\n#include \"HUSKYLENS.h\"\r\n\r\nHUSKYLENS huskylens;\r\n\/\/HUSKYLENS green line &gt;&gt; SDA; blue line &gt;&gt; SCL\r\nvoid printResult(HUSKYLENSResult result);\r\n\r\nvoid setup() {\r\n    Serial.begin(115200);\r\n    Wire.begin();\r\n    while (!huskylens.begin(Wire))\r\n    {\r\n        Serial.println(F(\"Begin failed!\"));\r\n        Serial.println(F(\"1.Please recheck the \\\"Protocol Type\\\" in HUSKYLENS (General Settings&gt;&gt;Protocol Type&gt;&gt;I2C)\"));\r\n        Serial.println(F(\"2.Please recheck the connection.\"));\r\n        delay(100);\r\n    }\r\n}\r\n\r\nvoid loop() {\r\n    if (!huskylens.request()) Serial.println(F(\"Fail to request data from HUSKYLENS, recheck the connection!\"));\r\n    else if(!huskylens.isLearned()) Serial.println(F(\"Nothing learned, press learn button on HUSKYLENS to learn one!\"));\r\n    else if(!huskylens.available()) Serial.println(F(\"No block or arrow appears on the screen!\"));\r\n    else\r\n    {\r\n        Serial.println(F(\"###########\"));\r\n        while (huskylens.available())\r\n        {\r\n            HUSKYLENSResult result = huskylens.read();\r\n            printResult(result);\r\n        }    \r\n    }\r\n}\r\n\r\nvoid printResult(HUSKYLENSResult result){\r\n    if (result.command == COMMAND_RETURN_BLOCK){\r\n        Serial.println(String()+F(\"Block:xCenter=\")+result.xCenter+F(\",yCenter=\")+result.yCenter+F(\",width=\")+result.width+F(\",height=\")+result.height+F(\",ID=\")+result.ID);\r\n    }\r\n    else if (result.command == COMMAND_RETURN_ARROW){\r\n        Serial.println(String()+F(\"Arrow:xOrigin=\")+result.xOrigin+F(\",yOrigin=\")+result.yOrigin+F(\",xTarget=\")+result.xTarget+F(\",yTarget=\")+result.yTarget+F(\",ID=\")+result.ID);\r\n    }\r\n    else{\r\n        Serial.println(\"Object unknown!\");\r\n    }\r\n}\r\n<\/pre>\n<p><span style=\"font-weight: 400;\">The only difference you\u2019ll note in the code is that we are using the Wire library instead of the SoftwareSerial library. The authors of the sketch didn\u2019t declare the Wire library, but as it is a built-in library, it isn\u2019t really necessary.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Note how the HuskyLens is bound to the wire object in its \u201cbegin\u201d statement.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Once again, you need to put your HuskyLens into the desired mode and load up the sketch.\u00a0 This time I did line tracking, and I used my finger as the line! In this mode, you get origin and target coordinates, as well as the line ID number. Great information if you\u2019re designing a line tracking robot.<\/span><\/p>\n<h2><span style=\"font-weight: 400;\">Modifying HuskyLens Text with Arduino<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">When recognizing faces or colors, it is a bit awkward to identify your targets by ID number. A proper name would be preferable.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">By writing data to the HuskyLens you can change the text displayed on the device&#8217;s screen.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Before we start the next experiment, you\u2019ll need to train the HuskyLens to recognize one or more objects or people.\u00a0 Once you do, you\u2019ll need to note the ID numbers assigned to each object or person.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">In my case, I trained it for four faces, which came up as ID numbers 1 through 4.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8290\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/modify-text-before.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/modify-text-before.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/modify-text-before.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">Now I\u2019ll run some code to put names to those faces.\u00a0 My HuskyLens is still connected using I2C.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Arduino Name-changing Sketch<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">The sketch to change the four ID numbers into names is as follows:<\/span><\/p>\n<pre class=\"lang:c++ decode:true \" title=\"huskylens-names.ino\">\/*\r\n  Assign names to detected HuskyLens faces\r\n  huskylens-names.ino\r\n\r\n  Uses DF Robot HuskyLens Library - https:\/\/github.com\/HuskyLens\/HUSKYLENSArduino\r\n\r\n  HuskyLens set for I2C interface\r\n\r\n  DroneBot Workshop 2021\r\n  <blockquote class=\"wp-embedded-content\" data-secret=\"UwWSHhB5y1\"><a href=\"https:\/\/dronebotworkshop.com\/\">Welcome to the Workshop!<\/a><\/blockquote><iframe class=\"wp-embedded-content\" sandbox=\"allow-scripts\" security=\"restricted\" style=\"position: absolute; clip: rect(1px, 1px, 1px, 1px);\" title=\"&#8220;Welcome to the Workshop!&#8221; &#8212; DroneBot Workshop\" src=\"https:\/\/dronebotworkshop.com\/embed\/#?secret=GDmbkqyXJs#?secret=UwWSHhB5y1\" data-secret=\"UwWSHhB5y1\" width=\"600\" height=\"338\" frameborder=\"0\" marginwidth=\"0\" marginheight=\"0\" scrolling=\"no\"><\/iframe>\r\n*\/\r\n\r\n\/\/ Include required libraries\r\n#include \"HUSKYLENS.h\"\r\n#include \"Wire.h\"\r\n\r\n\/\/ Create HuskyLens object\r\nHUSKYLENS huskylens;\r\n\r\nvoid setNewName (String newname, uint8_t ID) {\r\n  while (!huskylens.setCustomName(newname, ID))\r\n  {\r\n    Serial.println(F(\"Custom name failed!\"));\r\n    delay(100);\r\n  }\r\n\r\n}\r\n\r\nvoid setup()\r\n{\r\n\r\n  \/\/ Setup serial monitor\r\n  Serial.begin(115200);\r\n\r\n  \/\/ Connect to I2C bus as master\r\n  Wire.begin();\r\n\r\n  \/\/ Bind HuskyLens object to I2C conenction)\r\n  while (!huskylens.begin(Wire))\r\n  {\r\n    Serial.println(F(\"Begin failed!\"));\r\n    delay(100);\r\n  }\r\n\r\n  \/\/ Short delay to let HuskyLens stabilize\r\n  delay(100);\r\n\r\n  \/\/ Change detected face names\r\n  setNewName (\"John\", 1);\r\n  setNewName (\"Paul\", 2);\r\n  setNewName (\"George\", 3);\r\n  setNewName (\"Ringo\", 4);\r\n\r\n\r\n}\r\n\r\nvoid loop()\r\n{\r\n  \/\/ Nothing Here\r\n}\r\n<\/pre>\n<p><span style=\"font-weight: 400;\">We start by including both the HuskyLens and Wire libraries.\u00a0 We then define an object to represent our HuskyLens.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">I also defined a function to actually change the name, we\u2019ll get to it in a moment, but first we\u2019ll look at the Setup.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">In Setup, we start the Serial Monitor, which is only used for debug purposes. We also start the I2C connection and bind to the HuskyLens object.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">After a short delay to allow the HuskyLens to stabilize, we call the name changing function four times, once for each name we wish to change. Let\u2019s look at that function now.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The function, <\/span><i><span style=\"font-weight: 400;\">setNewName<\/span><\/i><span style=\"font-weight: 400;\">, simply calls an existing function that is part of the HuskyLens library. The HuskyLens<\/span><i><span style=\"font-weight: 400;\"> setCustomName<\/span><\/i><span style=\"font-weight: 400;\"> method does exactly what we want, it takes a name and an ID number as its input parameters.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">All that the function <\/span><i><span style=\"font-weight: 400;\">setNewName<\/span><\/i><span style=\"font-weight: 400;\"> does is call <\/span><i><span style=\"font-weight: 400;\">setCustomName <\/span><\/i><span style=\"font-weight: 400;\">and verify that it succeeded. If it didn\u2019t, then it is called again after a short delay.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">I found this to be more reliable than simply calling <\/span><i><span style=\"font-weight: 400;\">setCustomName<\/span><\/i><span style=\"font-weight: 400;\"> and adding a short delay, but you can experiment and see which one works for you.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">You\u2019ll probably want to modify the names I used in the sketch for your own situation. Once you do that, load the sketch to the Arduino while it is connected to the HuskyLens.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8289\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/modify-text-after.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/modify-text-after.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/modify-text-after.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">You should see the names change on the HuskyLens screen.<\/span><\/p>\n<h2><span style=\"font-weight: 400;\">Using the SD Card<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">The HuskyLens has a microSD card slot on its face, the card inserts with the contacts down.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">You can use the microSD card to save photos, although the resolution is limited to the 320&#215;240 size that you see on the HuskyLens screen.\u00a0 You can also save snapshots, which are photos with the screen text as well.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The SD card can also be used to save models you have learned. So, for example, if you teach the HuskyLens to recognize a face, you can save it to the SD card and then call it up later.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">To save a model, for example some faces you have learned, press down on the selector wheel so that you activate the bottom menu. Scroll to the end of the menu, where you will see an <\/span><i><span style=\"font-weight: 400;\">Export to SD<\/span><\/i><span style=\"font-weight: 400;\"> selection.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">When you select this, you\u2019ll get a scroll menu that goes from 0 to 5. This allows you to assign a number to your saved data, so that you can save more than one training set.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">After you have saved the data, you can recall it anytime. If you want to give that a try, then save some learned data and then go into General Settings and do a Factory Reset. This will wipe all the saved models from your HuskyLens.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-8270\" src=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/huskylens-factory-reset.png?resize=750%2C422&#038;ssl=1\" alt=\"\" width=\"750\" height=\"422\" srcset=\"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/huskylens-factory-reset.png?w=768&amp;ssl=1 768w, https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/huskylens-factory-reset.png?resize=300%2C169&amp;ssl=1 300w\" sizes=\"(max-width: 750px) 100vw, 750px\" data-recalc-dims=\"1\" \/><\/p>\n<p><span style=\"font-weight: 400;\">Then go back into Face Recognition (assuming that it was face data you saved) and activate the bottom menu. Scroll over to<\/span><i><span style=\"font-weight: 400;\"> Import from SD<\/span><\/i><span style=\"font-weight: 400;\"> and select it. You\u2019ll be prompted for the number of the saved settings, choose the same one you saved.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">After importing, you will need to reset the HuskyLens (power it off and back on). After that, try out the Face Recognition. You should see your saved data models are back now.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">You can save data for virtually all the HuskyLens modes.<\/span><\/p>\n<h2><span style=\"font-weight: 400;\">Conclusion<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">I was pretty impressed by the HuskyLens, this low-cost device really worked pretty well. It will certainly open up a world of new and exciting projects that can include machine vision.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">What is particularly fascinating about this device is that using it,\u00a0 you can start building AI projects without requiring any AI training.\u00a0\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">I\u2019ve actually started on one of those projects myself, a small robot based upon the HuskyLens 9and a few other technologies). I\u2019ll be presenting that soon, so keep a lookout for it here on the DroneBot Workshop website.<\/span><\/p>\n<div class=\"dbwspartsbox\">\n<h3>Parts List<\/h3>\n<p><em>Here are some components that you might need to complete the experiments in this article. Please note that some of these links may be affiliate links, and the DroneBot Workshop may receive a commission on your purchases. This does not increase the cost to you and is a method of supporting this ad-free website.<\/em><\/p>\n<p>COMING SOON!<\/p>\n<\/div>\n<p>&nbsp;<\/p>\n<div class=\"dbwsresourcebox\">\n<h3>Resources<\/h3>\n<p><a href=\"https:\/\/wiki.dfrobot.com\/HUSKYLENS_V1.0_SKU_SEN0305_SEN0336\" target=\"_blank\" rel=\"noopener\">HuskyLens Wiki<\/a> &#8211; DFRobot&#8217;s official HuskyLens Wiki.<\/p>\n<p><a href=\"https:\/\/s3.amazonaws.com\/download.dronebotworkshop.com\/HuskyLens+-+DroneBot+Workshop.zip\">Code for this article<\/a> &#8211; The code presented today in a ZIP file<\/p>\n<p><a href=\"https:\/\/github.com\/HuskyLens\" target=\"_blank\" rel=\"noopener\">HuskyLens GitHub Page<\/a> &#8211; The GitHub page supporting the HuskyLens<\/p>\n<p><a href=\"https:\/\/www.silabs.com\/developers\/usb-to-uart-bridge-vcp-drivers\" target=\"_blank\" rel=\"noopener\">Silicon Labs USB UART Bridge<\/a> &#8211; Download the UART Bridge for Windows<\/p>\n<\/div>\n<p>&nbsp;<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Today we will look at a powerful yet inexpensive machine vision sensor, the HuskyLens.<\/p>\n<p>This amazing little device is based upon the powerful Kendryte K210 processor and is capable of adding machine vision capabilities to your projects.<\/p>\n","protected":false},"author":1,"featured_media":8264,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"jetpack_post_was_ever_published":false,"_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":false,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[90],"tags":[],"jetpack_sharing_enabled":true,"jetpack_featured_media_url":"https:\/\/i0.wp.com\/dronebotworkshop.com\/wp-content\/uploads\/2021\/11\/HuskyLens.jpg?fit=768%2C432&ssl=1","jetpack_shortlink":"https:\/\/wp.me\/p74ClK-29W","jetpack-related-posts":[],"_links":{"self":[{"href":"https:\/\/dronebotworkshop.com\/wp-json\/wp\/v2\/posts\/8304"}],"collection":[{"href":"https:\/\/dronebotworkshop.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/dronebotworkshop.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/dronebotworkshop.com\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/dronebotworkshop.com\/wp-json\/wp\/v2\/comments?post=8304"}],"version-history":[{"count":7,"href":"https:\/\/dronebotworkshop.com\/wp-json\/wp\/v2\/posts\/8304\/revisions"}],"predecessor-version":[{"id":9842,"href":"https:\/\/dronebotworkshop.com\/wp-json\/wp\/v2\/posts\/8304\/revisions\/9842"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/dronebotworkshop.com\/wp-json\/wp\/v2\/media\/8264"}],"wp:attachment":[{"href":"https:\/\/dronebotworkshop.com\/wp-json\/wp\/v2\/media?parent=8304"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dronebotworkshop.com\/wp-json\/wp\/v2\/categories?post=8304"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dronebotworkshop.com\/wp-json\/wp\/v2\/tags?post=8304"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}